The Breakdown | Explaining Southern California's economy
Business & Economy

Good money after bad: CalPERS and the crisis in venture capital

The California Public Employees' Retirement System (CalPERS) building in Sacramento. The fund has billions invested in venture capital and other alternative asset classes.
The California Public Employees' Retirement System (CalPERS) building in Sacramento. The fund has billions invested in venture capital and other alternative asset classes.
Max Whittaker/Getty Images
The California Public Employees' Retirement System (CalPERS) building in Sacramento. The fund has billions invested in venture capital and other alternative asset classes.
The performance of VC has been poor sine the the mid-1990s.
The California Public Employees' Retirement System (CalPERS) building in Sacramento. The fund has billions invested in venture capital and other alternative asset classes.
In VC, smaller funds outperform larger ones.
The California Public Employees' Retirement System (CalPERS) building in Sacramento. The fund has billions invested in venture capital and other alternative asset classes.
Big VC doesn't meet outsized return expectations.
The California Public Employees' Retirement System (CalPERS) building in Sacramento. The fund has billions invested in venture capital and other alternative asset classes.
The "N-curve" is in blue and red. The "J-curve" is in green. In VC, you expect J, but you actually get N, with most returns coming early, rather than late.
The California Public Employees' Retirement System (CalPERS) building in Sacramento. The fund has billions invested in venture capital and other alternative asset classes.
A healthy VC ecosystem? Not really. Most of the money is going to fewer funds.
The California Public Employees' Retirement System (CalPERS) building in Sacramento. The fund has billions invested in venture capital and other alternative asset classes.
Are VCs really all about raising money and charging fees? The Kauffman Foundation blames itself for creating this compensation structure.

Pretend, for a moment, that you’re a computer science student at Stanford University. Chances are good that you’ve thought about taking your degree — or even not waiting to get your degree — and starting a technology company.

It’s the new American Dream. It attracts the most talented international students to our major research universities. It’s made the likes of Jerry Yang, Sergey Brin, Larry Page and, more recently, Facebook’s Mark Zuckerberg and Instagram’s Kevin Systrom (both under 30) multi-millionaires if not multi-billionaires nearly overnight.

Technology. The Internet. Mobile. Innovation. Disruption. Entrepreneurship.

These are the things that make America great in the early 21st century. Many of these new businesses are located in California. And they all have one thing in common: They live and die based on the investment decisions of venture capitalists, arguably the most important reallocators of wealth in the global economy.

Venture capital is the rocket fuel that gives scrappy tech startups in Silicon Valley, Los Angeles and San Diego (and other cities outside California) their liftoff velocity. Wanna be a tech tycoon? Then you’d better meet some VCs.

But where do VCs get their billions? For the most part, from foundations, endowments, pension funds and high-net worth individuals. And as the VC business has gotten tougher in the years since the bursting of the bubble on the late 1990s, those big funders have begun to ask some serious questions about what kind of return they’re getting on risky, long-term investments in the startup economy.

We Have Met the Enemy...And He Is Us

Last week, the Kauffman Foundation published a paper, jointly authored by Diane Mulcahy, Bill Weeks and Harold Bradley, titled "We Have Met the Enemy…And He Is Us: Lessons from Twenty Years of the Kauffman Foundation’s Investments in Venture Capital Funds and The Triumph of Hope over Experience."

It didn’t mince words. The conclusion was that Kauffman hasn’t just been investing unwisely in VC — it’s been perpetuating a myth that venture capital is a good way for investors to beat the public markets.

The foundation knows what it's talking about. "We have structured a compensation system that rewards fundraising," Mulcahy said of dynamic between VCs and funding sources, and of the way VCs are paid.

So VCs aren't creating the returns that they're supposed to. And you might see the headlines — “VC Doesn’t Deliver on Its Promises,” “VC Is a Bad Investment” — and conclude that VC has lost its mojo. That VCs, despite their professed expertise, kind of suck at finding the next Facebook. 

But Kauffman tells a different story, one of VCs and their funding sources locked into a dysfunctional relationship that’s actually preventing VCs from doing their jobs — and making it likely that funders will keep failing at theirs.

Bottom line: funders have put so much faith and money into VC that they’ve encouraged VCs to excel at…raising money, lots of money, not at finding great companies that will deliver market-beating returns. And the winnowing of VC over the past decade, with far fewer firms chasing many more dollars, has made the situation worse.

Kauffman is uniquely situated to study this issue. The foundation was started by healthcare tycoon Ewing Kauffman with a mission to promote and enable entrepreneurship. Based in Kansas City, Missouri, it's been around since the mid-1960s and has been very successful, with an endowment of nearly $2 billion.

But more importantly, it’s been investing in VC for 20 years and has amassed an enviable pile of data on this asset class.

A decent chunk of the Kauffman endowment — almost $250 million — is invested in a portfolio that's riskier than run-of-the-mill stocks and bonds. In order to garner higher returns, it's invested in such alternative investments as venture capital and private equity funds. The VC component is the focus of the Mulcahy-Weeks-Bradley paper, which draws on Kauffman's experience investing in innovative, entrepreneurial startups — quite literally putting its money where its mouth is, as Kauffman’s mission is to promote entrepreneurship.

Not What Everyone in VC Wanted to Hear

When the paper was released last week, it hit like a bombshell. Venture capitalists are already answering a lot of questions about the viability of their business model. The Kauffman paper added some serious fuel to that building firestorm of debate, which at its most apocalyptic has prompted numerous bloggers and commenters to declare that VC is "broken" or "dead," at least as it's been practiced since the 1990s.

Kauffman didn’t sugarcoat it. "Venture capital has delivered poor returns for more than a decade," the authors wrote. "VC returns haven’t significantly outperformed the public market since the late 1990s, and, since 1997, less cash has been returned to investors than has been invested in VC."

The response was swift and merciless. VC is "going to have to adjust to the new reality," wrote Stacey Higgenbotham at GigaOM. A "damning study," argued Russ Garland at the Wall Street Journal. "[I]nvestors who back venture capital would be better off trying their luck in public equity markets," said Reuters' James Saft.

However, the Kauffman authors weren't leveling the blame on VCs; they were placing it squarely on their own shoulders and, more broadly, those of "limited partners" (LPs) — the big foundations, endowments and state pension funds that have moved into VC in order to capture higher returns — everywhere. There are problems with "general partners" (GPs) — the VCs themselves — but those problems were inculcated by the culture that LPs encouraged.

The Big-Time World of Alternative Investments

So why have LPs invested in VC in the first place? For Kauffman, the answer is obvious: a foundation devoted to entrepreneurship needs to find a way to fund…entrepreneurs!

However, for big state pension funds — and in California, we have the biggest, CalPERS, at $231.9 billion currently — so-called "alternative investments," including VC, used to generate the loftier returns that these organizations need to meet their performance benchmarks, which in turn means that the funds can remain solvent, paying out benefits to retired state workers.

In recent years, those targets have been under plenty of stress. As large as CalPERS is, it was much larger before the financial crisis, at more than $260 billion. And in 2007, it was fully funded, but it's now only 70 percent funded. When I first wrote about CalPERS and its challenges, it had a 7.75 annual-return benchmark in order to meet its obligations. However, it's only ever beaten that if you calculate returns on a 20-year basis, with 8.38 percent. And since 2006, CalPERS has managed only 3.41 percent. Earlier this year, it dropped its benchmark for the entire fund to 7.50.

This recalibration is happening while many are rigorously questioning the commitment of state pension funds to alternative investments. The New York Times tackled the issue last month, comparing the performance of alternative investments — or, more accurately, lack of performance — in the pensions funds of Pennsylvania and Georgia.

Pennsylvania is far more committed, according to the Times: "The $26.3 billion Pennsylvania State Employees’ Retirement System has more than 46 percent of its assets in riskier alternatives, including nearly 400 private equity, venture capital and real estate funds."

Georgia, meanwhile, has a $14.4 million retirement fund that's prohibited from investing in alternative assets. And wouldn't you know it, on balance Georgia is doing better, with a 5.3 percent five-year return, against Pennsylvania's 3.6 percent (itself well below the 8 percent return target set by the state, which could explain why Pennsylvania has so much of its funds in the alternative category).

CalPERS currently has $50 billion in alternative investments, a mix of VC, private equity ("buyout," in the fund's language) and other riskier asset classes. Of that, about two-thirds, $34.2 billion, is fully funded, according to data that CalPERS provided me. The balance, $15.8 billion, was unfunded in June of last year, meaning that it’s money they anticipate having, but don’t yet. Of that $50 billion pie, about 58 percent is in private equity and 7 percent is in VC.

You can understand the imbalance: private equity has yielded a 13.9 percent return for CalPERS, while VC has managed only 2.7 percent, according to the Sacramento Business Journal. Those numbers have made the fund understandably skittish about expanding its exposure to VC. Private equity hasn't been as worrisome: according to Reuters, CalPERS just committed $500 million to the Blackstone Group, a huge firm with $190 billion in assets under management that just last year received $1.8 billion from a New Jersey pension fund. It's a trend: two other private-equity giants, KKR and Apollo Global Management, got $3 billion last month from the Teacher Retirement System of Texas.

Private equity — formerly known as "leveraged buyout" — is the business that created Mitt Romney's fortune. It isn't as sexy as VC. Starting innovative companies is far more exciting than buying up struggling ones as "distressed assets" using gobs of debt, then taking them private and banging them into profitable shape before selling them back to the public markets.

But the private equity exit strategy is more lucrative. To return capital to investors, VCs depend on the companies they fund either making it to initial public offerings, of the sort that Facebook is about to undertake, or else being acquired by a larger company. That's great when there's a robust IPO market. But right now, there isn't.

VC, We Have a Problem

As the Kauffman paper notes, VCs are stretching out their fund lifetimes longer than in the past (more than ten years is the New Normal). Meanwhile, legislation like the recently passed JOBS Act will enable startups to stall on their IPOs by increasing the threshold on the number of private shareholders (formerly 500) a company can have before the SEC requires it to report financials to the Securities and Exchange Commission (SEC).

Even a company like Facebook, with perhaps the most hotly anticipated IPO of all time, would prefer to remain private, but can't because it ran up against the SEC private-shareholder limit. To deal with this, Facebook is selling only about 10 percent of the company, looking to raise enough money to allow its biggest funders to exit. It's also set up the way its employees are paid in stock to avoid them actually owning shares, prior to the IPO. But if CEO and co-founder Mark Zuckerberg had a choice, he'd keep Facebook private. Forever.

This all adds up to funds and foundations, like Kauffman and CalPERS, waiting longer to see their trapped VC investments become "liquid,” and thus easily sold. And in any case, as the Kauffman authors wrote, even when companies exit and VCs can return investments to their LPs, those returns have been negative for more than a decade.

And there's another troubling phenomenon at play, one that's provoked comment from several prominent VCs: From 1985 to 1995, VC was dealing with $500 million a year. Since then, it's been trying to deal with $20 billion annually — a daunting 40-fold increase.

Hold on, you might say. Aren’t there thousands of startups out there, eager for funding? Well, yes and no. There are lots of entrepreneurs who have an idea for a product or a service. But do they have a business model? In many cases, no. And the last thing VCs want to do is fund companies just because they exist, or because they might have some technologies or patents that they can port over to another company.

But more money means more pressure to put that money to work, which in turns creates a classic bubble: too many dollars chasing too few good opportunities, with all the bad bets that make millions disappear further undermining VC’s reputation.

That big $20 billion number that’s put into venture capital might also be underestimating the scale of yearly VC funding. In a recent talk at Grind, a New York co-working space, Fred Wilson of Union Square Ventures said the figure was actually $30 billion — and indicated that, through crowdfunding, the number could rise to $300 billion!

In a continuation of the reaction to the Kauffman paper, Wilson's comments were seized by, among others, Forbes, as evidence that the VC business model is broken, perhaps even dead. Wilson, on his blog, reacted forcefully to that dire prognosis: "The venture capital business is not dying," he wrote. "This post is a plea to bloggers and journalists not to use the word 'death' casually. It is a big word, a strong word, it means something real and devastating. And it is a word I would not use lightly."

Wilson has been in the forefront on outlining a new model for VC, potentially addressing some of the problems highlighted in the Kauffman paper. He's mentioned in the paper as being an advocate for greater transparency in VC fund performance. He's suggested that VC might split into two factions, with one side playing by the old rules, while a new, more nimble counterpoint emerges that sees VC firms acting more like innovation consultants, offering less money but more mentoring and advice, in exchange for equity.

I contacted USV for this story, but Wilson declined to comment, as did his partners.

However, I was able to talk with Kauffman's Mulcahy — herself a VC veteran, prior to joining Kauffman as Director of Private Equity — who addressed the question of all the money coming into VC and how large pension funds deal with it. "Pension funds feel a lot pressure to invest big amounts," she said. "So they look to big [VC] funds to put bigger amounts of money to work."

Why It's Hip to Be Small

However, in an additional negative wrinkle, she and her co-authors discovered that for the Kauffman Foundation, smaller VC funds have performed better than the big boys. "No fund greater than $1 billion has returned more than twice the capital invested, net of fees. The best-performing funds are consistently smaller funds."

That makes it simple to figure out how to fix the problem of VC in alternative investment portfolios, right? Just concentrate on small funds.

But that's easier said than done. For starters, smaller funds may not be prepared to deal with large capital inflows — they have to find good startups to invest in, after all, and while they may not put all of a fund to work right away, they don't necessarily want to sit on their money. So they can have too much in the bank.

They may be able to solve this problem by becoming “followers rather than leaders.” David Siemer of Siemer Ventures in Los Angeles told me that this is how his $35 million fund prefers to operate: joining with other VCs to invest in early-stage companies, taking some of the pressure off, letting others assume the risk of finding the next big thing.

But even then, it’s not that simple. As I noted in early April, a winner-take-all dynamic is emerging in VC:

[I]n 2012… [VC] fundraising is looking up, with a first quarter total of almost $4.9 billion. 

This is a pace that could deliver a $20-billion year in 2012.

But there's some bad news. Only 42 VC funds participated in these raises, waaayyy down from the 212 funds that were raising money in 2008. Also, a winner-take-all mentality is taking hold, decisively, in VC as five funds have accounted for 75 percent of total first-quarter fundraising — $3.6 billion. That leaves 37 other funds fighting over just over a billion in remaining funding. 

Andreessen Horowitz alone accounted for $1.5 billion, meaning that one VC firm gobbled up 42 percent of funding that was raised.

So even though, in Kauffman's case, it's the sub-$500-million funds that have shown the best returns, the VC ecosystem is functioning with fewer and fewer funds, and it’s the heavy hitters who are dominating fundraising (the Kauffman paper labels this the "Institutionalization of Venture Capital").

The Myth of the "J-curve"

And there's another factor to consider: the way that VC funds are technically performing also seems to be shifting. Mulcahy broke down the so-called "J-curve" for me, which she and her co-authors characterized as a "myth."

"Early on, investors should expect a negative return," she said. (This is the downcurve in the "J.") She continued: "The fund will call down fees and write off early 'lemon' investments, and this will persist for the first several years" of the fund's life. "The best returns will then be harvested at the end."

But something completely different happened with Kauffman's VC funds — and according to publicly available data, also happened with CalPERS (as well as CalSTRS, which isn't as enormous as CalPERS, but is still the eleventh-largest pension fund in the world). "[R]ather than showing weak early IRRs (a J-curve), CALPERS fund data for the period show that more than twice as many fund IRRs are positive than negative. CALSTRS reports similar data." ("IRR" stands for "internal rate of return," which is basically a way of estimating how much an investment can be expected to grow over time.)

It gets worse: "These data suggest that the J-curve effect is mostly notable by its absence. More distressingly, it suggests that too many fund managers focus on the front end of a fund’s performance period because that performance drives a successful fundraising outcome in subsequent funds."

Kauffman identified what it called an "N-curve" for its VC investment, with positive returns occurring early in the life of VC funds. Returns spike early, then trail off. This makes it possible for VCs to claim early success, thereby enabling them to bring in more money in subsequent raises, while failing to make good on their ostensible business model — represented by the J-curve — to lose money before they deliver outsize returns much later. Kauffman maintains that LPs have bought into this arrangement with GPs, bolstering the myth of the J-curve, which has cemented the infamous "2 and 20" compensation structure for VCs, with managers receiving a 2 percent management fee and 20 percent of the eventual profits.

Obviously, if the return targets are rarely met, then there's a perverse incentive to keep funds going, so that management fees can be extracted. In the paper, Kauffman argued that this isn't the fault of nefarious GPs, but rather LPs who have boxed themselves into return goals that force them to perpetuate a 2-and-20 system that isn't working.

In other words, VCs are getting much better at asking for — and getting — money than they are at investing in companies that will, through the alchemy of innovation and rapid growth, transform that money into massive returns. 

And what about using VC to prevent big pension funds from falling behind on their return targets? "VC is a really hard way to play catch up," she said. Unfortunately, many institutions may not want to alter this structure, for fear of not having access to the largest and most well known VCs, even though those firms might not be delivering on expectations.

Mulcahy argued that moving away from IRR as a measure might offer a way out of this bind. "IRR is just a bad measure. It's fairly easy to manipulate." She and her co-authors recommended using a "public market equivalent" (PME) to benchmark VC fund performance. "A PME is informative. It tells us if a VC portfolio is outperforming the public market." Kauffman suggested the Russell 2000, an index of small-cap stocks that approximates the financial dynamics of startup companies.

The CalPERS Point of View

On the CalPERS front, spokesperson Brad Pacheco provided me with a presentation from the fund's most recent annual review that showed benchmarking against both the Wilshire 2500 (minus tobacco stocks) and a "policy" benchmark developed using a customized version of the FTSE All-World Index, which is an aggregate of 2800 large- and mid-cap global stocks. The data were for the entire alternative investment portfolio, not just the VC piece. They showed CalPERS alternative investments actually beating the Wilshire benchmark, but "lagging" the policy benchmark for three- and five-year time frames.

"We have a relatively large investment in 300 funds," Pacheco said. "And we're currently going through a strategic alignment because we have such a large grouping. But we haven't changed the asset allocation in quite a while."

The important thing to note here is that Kauffman is a fraction of the size of CalPERS. CalPERS has billions more invested in just the VC component of its alternative portfolio than Kauffman has in its entire endowment. So it's hard to make an apples-to-apples comparison.

However, CalPERS does appear to be suffering from the mistaken allegiance to the myth of the J-curve that Kauffman accused itself of. The question is whether both organizations are, in the words of the Kauffman authors, "succumb[ing] time and again to narrative fallacies, a well-studied behavioral finance bias."

Kauffman believes that it is and has proposed to make major changes to the way it deals with VC in the future. CalPERS, at more than 100 times Kauffman's size, may move a bit more slowly. But returns can be relative. And if Kauffman is right in its analysis, then it may start seeing the kind of benchmark-beating returns that VC promises much earlier than CalPERS does.

So what if you are that Stanford student, hoping to hit it big? You might need to deal with the prospect of having a lot less money than those who preceded you in the quest for the American Dream. And while that might sound bad, in actuality it could be the best thing that ever happened to that company you haven’t yet started to produce a product that you haven’t yet invented. Instead of worrying about making someone else a bunch of money, you can worry about building a solid business down the road.

Follow Matthew DeBord and the DeBord Report on Twitter.